How to read a scientific abstract

[drwex]

http://www.huffingtonpost.com/noah-gray/abstract-science_b_1923214.html

Over at the HuffPo, Noah Gray (who is a senior editor at Nature) has a really good step-by-step "how to" on reading abstracts of scientific papers. Often with prepublications, or authors' Web sites or citation sources like SSRN the abstract is all you have to go by and it's nice when an author makes the abstract clear. Gray breaks down what a (well constructed) abstract looks like.

In particular, it's often easy from an abstract to see exactly where mainstream media has gone off the rails in reporting. Though I do wish more scientists would put a little more into the well portion of "well-written abstract."

(h/t Maggie Koerth-Baker at Boingboing for the original pointer)

Comments

[selenite.livejournal.com]

I tend to skim until I find something like "we observed a sample of 36 children" when researching the cause of a disease with a 1 in 200 (or 2000) incidence rate. Then bit bucket it. I've gotten very down on the current state of scientific research.

[drwex]

Get in line :)

I spent much of my time in the CHI community complaining about the tiny sample sizes of users in published papers.

But, the point is still good - when I read a paper i usually jump from the abstract to the methodology section.